# The Grammar-Learning Trajectories of Neural Language Models

arXiv: https://arxiv.org/abs/2109.06096 · HTML: https://mlsanity.com/paper/2109.06096/

## Key facts

- Cited by 36 papers in the map (data through 14 Oct 2026).
- Builds on 31 of its 67 references that are in the map.
- First posted on arXiv 13 Sep 2021.

## Builds on (most cited first, 30 of 31)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Transformer-XL: Attentive Language Models Beyond a Fixed-Length Context](https://mlsanity.com/paper/1901.02860/) (2019)
- [Aligning Books and Movies: Towards Story-like Visual Explanations by Watching Movies and Reading Books](https://mlsanity.com/paper/1506.06724/) (2015)
- [Right for the Wrong Reasons: Diagnosing Syntactic Heuristics in Natural Language Inference](https://mlsanity.com/paper/1902.01007/) (2019)
- [Analyzing Multi-Head Self-Attention: Specialized Heads Do the Heavy Lifting, the Rest Can Be Pruned](https://mlsanity.com/paper/1905.09418/) (2019)
- [Quantifying Attention Flow in Transformers](https://mlsanity.com/paper/2005.00928/) (2020)
- [An Empirical Study of Example Forgetting during Deep Neural Network Learning](https://mlsanity.com/paper/1812.05159/) (2018)
- [BLiMP: The Benchmark of Linguistic Minimal Pairs for English](https://mlsanity.com/paper/1912.00582/) (2019)
- [What do you learn from context? Probing for sentence structure in contextualized word representations](https://mlsanity.com/paper/1905.06316/) (2019)
- [Dataset Cartography: Mapping and Diagnosing Datasets with Training Dynamics](https://mlsanity.com/paper/2009.10795/) (2020)
- [On The Power of Curriculum Learning in Training Deep Networks](https://mlsanity.com/paper/1904.03626/) (2019)
- [Evaluating Gender Bias in Machine Translation](https://mlsanity.com/paper/1906.00591/) (2019)
- [The GEM Benchmark: Natural Language Generation, its Evaluation and Metrics](https://mlsanity.com/paper/2102.01672/) (2021)
- [WikiMatrix: Mining 135M Parallel Sentences in 1620 Language Pairs from Wikipedia](https://mlsanity.com/paper/1907.05791/) (2019)
- [Identifying and Controlling Important Neurons in Neural Machine Translation](https://mlsanity.com/paper/1811.01157/) (2018)
- [BERTs of a feather do not generalize together: Large variability in generalization across models with similar test set performance](https://mlsanity.com/paper/1911.02969/) (2019)
- [A Theoretical Analysis of the Repetition Problem in Text Generation](https://mlsanity.com/paper/2012.14660/) (2020)
- [A Walk with SGD](https://mlsanity.com/paper/1802.08770/) (2018)
- [Probing Across Time: What Does RoBERTa Know and When?](https://mlsanity.com/paper/2104.07885/) (2021)
- [Making Transformers Solve Compositional Tasks](https://mlsanity.com/paper/2108.04378/) (2021)
- [Let's Agree to Agree: Neural Networks Share Classification Order on Real Datasets](https://mlsanity.com/paper/1905.10854/) (2019)
- [What Do Position Embeddings Learn? An Empirical Study of Pre-Trained Language Model Positional Encoding](https://mlsanity.com/paper/2010.04903/) (2020)
- [Recurrent Neural Networks in Linguistic Theory: Revisiting Pinker and Prince (1988) and the Past Tense Debate](https://mlsanity.com/paper/1807.04783/) (2018)
- [Quantity doesn't buy quality syntax with neural language models](https://mlsanity.com/paper/1909.00111/) (2019)
- [Easy over Hard: A Case Study on Deep Learning](https://mlsanity.com/paper/1703.00133/) (2017)
- [Inherent Biases in Reference based Evaluation for Grammatical Error Correction and Text Simplification](https://mlsanity.com/paper/1804.11254/) (2018)
- [Automatically Extracting Challenge Sets for Non local Phenomena in Neural Machine Translation](https://mlsanity.com/paper/1909.06814/) (2019)
- [Mediators in Determining what Processing BERT Performs First](https://mlsanity.com/paper/2104.06400/) (2021)
- [Classifying Syntactic Errors in Learner Language](https://mlsanity.com/paper/2010.11032/) (2020)

## Built on by (most cited first, 30 of 36)

- [Memorization Without Overfitting: Analyzing the Training Dynamics of Large Language Models](https://mlsanity.com/paper/2205.10770/) (2022)
- [Active Learning on a Budget: Opposite Strategies Suit High and Low Budgets](https://mlsanity.com/paper/2202.02794/) (2022)
- [Rho-1: Not All Tokens Are What You Need](https://mlsanity.com/paper/2404.07965/) (2024)
- [Sudden Drops in the Loss: Syntax Acquisition, Phase Transitions, and Simplicity Bias in MLMs](https://mlsanity.com/paper/2309.07311/) (2023)
- [Fusing finetuned models for better pretraining](https://mlsanity.com/paper/2204.03044/) (2022)
- [What Artificial Neural Networks Can Tell Us About Human Language Acquisition](https://mlsanity.com/paper/2208.07998/) (2022)
- [Language Model Behavior: A Comprehensive Survey](https://mlsanity.com/paper/2303.11504/) (2023)
- [Training Trajectories of Language Models Across Scales](https://mlsanity.com/paper/2212.09803/) (2022)
- [LLM Circuit Analyses Are Consistent Across Training and Scale](https://mlsanity.com/paper/2407.10827/) (2024)
- [ColD Fusion: Collaborative Descent for Distributed Multitask Finetuning](https://mlsanity.com/paper/2212.01378/) (2022)
- [Characterizing Learning Curves During Language Model Pre-Training: Learning, Forgetting, and Stability](https://mlsanity.com/paper/2308.15419/) (2023)
- [Analyzing the Mono- and Cross-Lingual Pretraining Dynamics of Multilingual Language Models](https://mlsanity.com/paper/2205.11758/) (2022)
- [Neural Networks Learn Statistics of Increasing Complexity](https://mlsanity.com/paper/2402.04362/) (2024)
- [A Hitchhiker's Guide to Scaling Law Estimation](https://mlsanity.com/paper/2410.11840/) (2024)
- [RuCoLA: Russian Corpus of Linguistic Acceptability](https://mlsanity.com/paper/2210.12814/) (2022)
- [Understanding Transformers via N-gram Statistics](https://mlsanity.com/paper/2407.12034/) (2024)
- [A distributional simplicity bias in the learning dynamics of transformers](https://mlsanity.com/paper/2410.19637/) (2024)
- [DecoderLens: Layerwise Interpretation of Encoder-Decoder Transformers](https://mlsanity.com/paper/2310.03686/) (2023)
- [How to Select Which Active Learning Strategy is Best Suited for Your Specific Problem and Budget](https://mlsanity.com/paper/2306.03543/) (2023)
- [Latent State Models of Training Dynamics](https://mlsanity.com/paper/2308.09543/) (2023)
- [A Systematic Assessment of Language Models with Linguistic Minimal Pairs in Chinese](https://mlsanity.com/paper/2411.06096/) (2024)
- [Language Model Behavioral Phases are Consistent Across Architecture, Training Data, and Scale](https://mlsanity.com/paper/2510.24963/) (2025)
- [Evolution of Concepts in Language Model Pre-Training](https://mlsanity.com/paper/2509.17196/) (2025)
- [Predicting the Susceptibility of Examples to Catastrophic Forgetting](https://mlsanity.com/paper/2406.09935/) (2024)
- [What Happens During the Loss Plateau? Understanding Abrupt Learning in Transformers](https://mlsanity.com/paper/2506.13688/) (2025)
- [Large Language Models as Psychological Simulators: A Methodological Guide](https://mlsanity.com/paper/2506.16702/) (2025)
- [ErrorMap and ErrorAtlas: Charting the Failure Landscape of Large Language Models](https://mlsanity.com/paper/2601.15812/) (2026)
- [Bigram Subnetworks: Mapping to Next Tokens in Transformer Language Models](https://mlsanity.com/paper/2504.15471/) (2025)
- [Unforgettable Generalization in Language Models](https://mlsanity.com/paper/2409.02228/) (2024)
- [Preferences for Idiomatic Language are Acquired Slowly -- and Forgotten Quickly: A Case Study on Swedish](https://mlsanity.com/paper/2602.03484/) (2026)

## Abstract (arXiv)

The learning trajectories of linguistic phenomena in humans provide insight into linguistic representation, beyond what can be gleaned from inspecting the behavior of an adult speaker. To apply a similar approach to analyze neural language models (NLM), it is first necessary to establish that different models are similar enough in the generalizations they make. In this paper, we show that NLMs with different initialization, architecture, and training data acquire linguistic phenomena in a similar order, despite their different end performance. These findings suggest that there is some mutual inductive bias that underlies these models' learning of linguistic phenomena. Taking inspiration from psycholinguistics, we argue that studying this inductive bias is an opportunity to study the linguistic representation implicit in NLMs. Leveraging these findings, we compare the relative performance on different phenomena at varying learning stages with simpler reference models. Results suggest that NLMs exhibit consistent "developmental" stages. Moreover, we find the learning trajectory to be approximately one-dimensional: given an NLM with a certain overall performance, it is possible to predict what linguistic generalizations it has already acquired. Initial analysis of these stages presents phenomena clusters (notably morphological ones), whose performance progresses in unison, suggesting a potential link between the generalizations behind them.